Last modified: 2016-06-02
Abstract
The casting process includes the transportation process and the pouring process. The transportation process is to carry the molten metal from the place where it is saved to the mold or the injection molding machine, and the pouring process is to pour and fill molten metal in the mold.
In the transportation process, it is undesirable to expose the surface of the molten metal to air to suppress the defect (for instance, generation of the oxidizing layer). Therefore, it is necessary to move to pour it into the mold as soon as possible after the molten metal is bailed out by the ladle. However, it is also necessary to prevent overflow of the molten metal and intervention of air or gas. So the liquid vibration by the acceleration and deceleration of the transfer machine should be suppressed.
Then, the development of the technique to control the liquid surface oscillation is needed. Computational fluid dynamics is expected to be effective as the means of the verification. The phenomenon handled here should treat the wall in the container to be a moving boundary, and the liquid surface as a free boundary.
In the pouring process, the flow of the molten metal in the mold is the target of interest. Because the quality of a final article of cast is dependent on how the molten metal flows in complex shape of the mold. Also as for this process, the use of numeric fluid analysis on the process including the fission and the fusion on a free surface is expected to be effective.
The particle-based fluid analysis methods are considered as the numerical computation technique which is applicable and useful in the treatment of these moving boundary and free boundary. However, quantitative comparative studies with the data of actual transportation and pouring process are few.
We have applied the smoothed particle hydrodynamics method to the process of transportation and pouring and validated the results with experimental data. We report on the technique and the result because we saw the experimental data and our numerical results are a good agreement.